In recent years, IETF (Internet Engineering Task Force) has been considering an IP mobility scheme for seamless movement, which enables handover between a plurality of different wireless communication networks such as a cellular phone network, a wireless LAN and the like, in order to achieve ubiquitous environment. As a specific protocol of the IP mobility scheme, there are Mobile IPv4 and Mobile IPv6 (which are abbreviated as Mobile IP, hereinafter) for supporting movements of each individual communication terminal, and NEMO (Network Mobility) for supporting mobility of a network as a unit.
Incidentally, when an application such as VoIP having a real-time property is executed via the wireless communication network, an allowable bandwidth of a wireless communication path changes depending on a propagation environment such as fading, and arrival intervals of packets received by the communication terminal are changed in accordance with a change of the allowable bandwidth.
For this reason, it is generally performed to provide the communication terminal with a jitter buffer so as to first store received packets in the jitter buffer and then read out the packets from the jitter buffer and reproduce the packets at intervals based on the application. Thereby, it absorbs deviation in the packets, that is, displacement of reproduction intervals of packets caused by displacement of the arrival intervals (jitter) of the packets, so as to prevent deterioration of reproduction quality such as reproduced sound quality and the like. Moreover, when there is no packet in the jitter buffer since the jitter is large, and therefore silence occurs, or when too many packets are received in a short period to be stored in the jitter buffer, the communication terminal changes a reproduction speed, discards received packets, or changes a size of the jitter buffer.
On the other hand, a downlink absolute delay time of a packet received by the communication terminal, that is, a time (delay time) required for a packet transmitted from a counterpart communication terminal to be received via the wireless communication network varies depending on wireless communication networks. Thus, if the communication terminal is a wireless communication apparatus which is moving and performs handover to a different wireless communication network, and when the downlink absolute delay time of a handover destination is longer than that of a handover source, for example, it causes blank period not receiving a packet according to the a difference between the downlink absolute delay times.
In such a case, if the blank period not receiving the packet is longer than a time required to read out a last packet (that is, a jitter buffer standard delay time), received from the wireless communication network of the handover source, from the jitter buffer when packets are read out from the jitter buffer at certain intervals based on an application and reproduced at a certain reproduction speed, for example, there is no packet in the jitter buffer for a period of such exceeding time. As a result, since reproduction of packets is not performed at least during this period, it causes silence and deteriorates reproduction quality.
FIG. 12 shows diagrams for explaining a control method of the jitter buffer in the above case. In FIG. 12, (a) shows the number of packets received by the jitter buffer for a unit period, (b) shows the reproduction speed (read-out intervals) of packets from the jitter buffer, and (c) shows the number of packets in the jitter buffer. FIG. 13 shows flows of the packets in that case. In FIG. 13, “Transmission”, “Reception” and “Reproduction” represent a transmission timing of a packet by the counterpart communication terminal, a reception timing of the packet received by the jitter buffer of the wireless communication apparatus, and a reproduction timing of the packet (timing to read out the packet from the jitter buffer) by the wireless communication apparatus, respectively. Here, it is assumed that there is no deviation in received packets (displacement of arrival intervals) either at a wireless communication network A of the handover source and a wireless communication network B of the handover destination.
As obvious from FIG. 12 and FIG. 13, no packet is reproduced for a time Tn={(TddnB−TddnA)−Ta}, if a downlink absolute delay time TddnB of the wireless communication network B of the handover destination is longer than a downlink absolute delay time TddnA of the wireless communication network A of the handover source and the difference (TddnB−TddnA) is longer than a jitter buffer standard delay time Ta, which is applied to received packets when there is a standard number of packets in the jitter buffer. In such a case, moreover, since packets are reproduced immediately after being received from the wireless communication network B of the handover destination, it is not possible to absorb jitter.
In order to improve such a defect at handover, there is suggested a method, for example, to monitor a reception condition of packets and, if packets are not received at standard reception intervals, to control reading out of the packets from the jitter buffer, that is, to control the reproduction speed of the packets (see Patent Document 1, for example).
FIG. 14 shows diagrams illustrating a control method of the jitter buffer disclosed in Patent Document 1. In FIG. 14, in the same manner as FIGS. 12(a) to (c), (a), (b) and (c) show the number of packets received by the jitter buffer in a unit time, the reproduction speed, and the number of packets in the jitter buffer, respectively. FIG. 15 shows flows of the packets in such a case.
As shown in FIG. 14 and FIG. 15, if packets cannot be received at previous reception intervals such as when handover is performed from the wireless communication network A with the downlink absolute delay time TddnA to the wireless communication network B with the downlink absolute delay time TddnB which is longer than TddnA, the reproduction speed of packets in the jitter buffer is gradually reduced in accordance with increase in the reception intervals. When the reception intervals return to normal intervals thereafter, the reproduction speed is controlled to gradually increase to the standard reproduction speed in accordance with the number of packets in the jitter buffer.
Patent Document 1: Japanese Patent Laid-Open No. 2006-238445